agentscope.realtimeDashScopeAudioRealtimeModel
class DashScopeAudioRealtimeModel(DashScopeRealtimeModel)agentscope/realtime/_dashscope/_audio_model.py:16–282DashScope 音频实时模型，处理低延迟的实时语音与多模态交互。位于 agentscope.realtime 包，继承自 DashScopeRealtimeModel。核心成员包括 list_models、push_text、replay_history、request_response、input_transcription_enabled。
Same endpoint and framing as Qwen-Omni, but it accepts text turns,
spells both PCM formats as "pcm", and offers smart_turn
endpointing in place of semantic_vad.
嵌套类摘要Nested Class Summary
| 类Class | 类型Kind | 说明Description |
|---|---|---|
Parameters | 类class | 参数，处理低延迟的实时语音与多模态交互。 Tuneables surfaced to the UI. |
属性摘要Property Summary
| 属性Property | 说明Description |
|---|---|
input_transcription_enabled | 读取 `input_transcription_enabled` 属性。 Qwen Audio does not emit settled input transcription. |
属性摘要Attribute Summary
| 属性Attribute | 类型Type | 说明Description | 值Value |
|---|---|---|---|
type | — | 类型字段。 | 'dashscope_audio_realtime' |
supports_text_input | — | 支持文本输入字段。 | True |
supports_history_replay | — | `supports_history_replay` 字段。 | True |
parameters | 'DashScopeAudioRealtimeModel.Parameters' | 参数字段。 | — |
方法摘要Method Summary
| 方法Method | 说明Description |
|---|---|
def list_models(custom_yaml_dir: str | None = None) -> list[RealtimeModelCard] | 列出模型。 List the Qwen-Audio cards, which live beside the Omni ones. |
async def push_text(text: str) -> Noneasync | 设置推送文本。 Send a text user turn and ask for the reply. |
async def replay_history(messages: Sequence[Msg]) -> Noneasync | 设置该参数。 Insert completed history without asking the model to respond. |
async def request_response() -> Noneasync | 返回请求响应。 Ask for a reply with the Qwen-Audio modality override. |
属性详情Attribute Detail
type
= 'dashscope_audio_realtime'类型字段。
The type field.
agentscope/realtime/_dashscope/_audio_model.py:74supports_text_input
= True支持文本输入字段。
The supports text input field.
agentscope/realtime/_dashscope/_audio_model.py:75supports_history_replay
= Truesupports_history_replay 字段。
The supports history replay field.
agentscope/realtime/_dashscope/_audio_model.py:76parameters
'DashScopeAudioRealtimeModel.Parameters'参数字段。声明为 'DashScopeAudioRealtimeModel.Parameters'。
The parameters field.
agentscope/realtime/_dashscope/_audio_model.py:78属性详情Property Detail
input_transcription_enabled
def input_transcription_enabled() -> bool读取 input_transcription_enabled 属性。
Qwen Audio does not emit settled input transcription.
agentscope/realtime/_dashscope/_audio_model.py:81–83方法详情Method Detail
list_models
def list_models(custom_yaml_dir: str | None = None) -> list[RealtimeModelCard]列出模型。
List the Qwen-Audio cards, which live beside the Omni ones.
| 名称Name | 类型Type | 说明Description |
|---|---|---|
custom_yaml_dir默认default None | str | None | 自定义 yaml 目录（类型 `str | None`）。The custom yaml dir. |
agentscope/realtime/_dashscope/_audio_model.py:86–93push_text
async def push_text(text: str) -> None设置推送文本。
Send a text user turn and ask for the reply.
| 名称Name | 类型Type | 说明Description |
|---|---|---|
text | str | 文本（类型 `str`）。The text. |
agentscope/realtime/_dashscope/_audio_model.py:95–107replay_history
async def replay_history(messages: Sequence[Msg]) -> None设置该参数。
Insert completed history without asking the model to respond.
| 名称Name | 类型Type | 说明Description |
|---|---|---|
messages | Sequence[Msg] | 消息列表（类型 `Sequence[Msg]`）。The messages. |
agentscope/realtime/_dashscope/_audio_model.py:109–213request_response
async def request_response() -> None返回请求响应。
Ask for a reply with the Qwen-Audio modality override.
agentscope/realtime/_dashscope/_audio_model.py:215–224嵌套类详情Nested Class Detail
Parameters
class Parameters参数，处理低延迟的实时语音与多模态交互。位于 agentscope.realtime 包，继承自 RealtimeModelBase.Parameters。
Tuneables surfaced to the UI. Declared in agentscope.realtime. Extends RealtimeModelBase.Parameters.